KR20060126817A - Pre-distortion method, measurement arrangement, pre-distorter structure, transmitter, receiver and connecting device - Google Patents

Abstract

A transmitter and other features of a communication system include means for determining power of a signal to be pre-distorted for quantizing the power in a predetermined way and for selecting a corresponding power interval. The transmitter and other features also include means for deriving parameters defining a linear curve segment of the selected power interval are disclosed. The transmitter and other features also include means for calculating a gain factor by using the parameters defining the linear curve segment of the selected power interval and means for obtaining a pre-distorted signal.

Description

Predistortion method, measurement arrangement, predistorter structure, transmitter, receiver and connecting device

The present invention relates to a predistortion method, a predistorter measurement device, a predistorter structure, a receiver, a transmitter, and a connection device.

Due to the non-linear effects of the analog components of the transmission chain, the amplitude and phase of the transmission signal are disturbed. Such distortions depend on the signal magnitude. If the signal size is large, the distortions become quite large.

The main cause of non-linearity is the power amplifier of the transmitter. In addition to amplifying the desired signal, the power amplifier generates higher harmonics of the original signal spectrum. Spreading of the signal spectrum has two major effects that make the requirements for the radio frequency spectrum not met and cause the detection of a distortion signal at the receiver to face errors.

Spreading of the signal spectrum can be avoided by reducing the signal power. However, this leads to inefficient use of the amplifier stage.

It is an object of the present invention to provide an improved predistortion method, predistorter measurement device, predistorter structure, receiver, transmitter, and connection device.

According to one aspect of the present invention, there is provided a predistortion method in a communication system, wherein the communication system includes at least one transmitter and at least one receiver, wherein each predistortion method comprises: Defining symbol and quadrature differences of the received signal sample and symbol determination for the point of transition, and transmitting the in-phase and quadrature differences defined above to the transmitter, forming power intervals based on the modulation method used, predistortion Determining the power of the signal to be quantized, quantizing the power in a predetermined manner and selecting a corresponding power interval, averaging the in-phase and orthogonal differences and averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes And perform amplification and moving average calculations between the selected powers. Linear curve and finding the parameters that define the segment of, and by calculating the gain factor from the parameters defining the linear curve segment of the selected power interval comprises the step of obtaining a pre-distortion signal.

According to another aspect of the present invention, there is provided a predistortion method in a communication system, wherein the communication system includes at least one transmitter and at least one receiver, wherein the predistortion method comprises a respective cone. Defining symbol and quadrature differences of the received signal sample and symbol determination for the stealation point, and transmitting the in-phase and orthogonal differences defined above to the transmitter, forming power intervals based on the modulation method used, advance Determining the power of the signal to be distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval, obtaining parameters defining a linear curve segment of the selected power interval, and the selected power interval Gain through parameters defining the linear curve segment of Calculated cut characterized in that it comprises a step of obtaining a pre-distortion signal.

According to another aspect of the invention, there is provided a receiver of a communication system, the receiver comprising means for defining in-phase and quadrature differences of symbol determination and received signal samples for each constellation point, and the in-phase defined above. And means for transmitting orthogonal differences to the transmitter.

According to another aspect of the present invention, there is provided a transmitter of a communication system, the transmitter comprising means for determining the power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Average the in-phase and quadrature differences, rotate the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and perform amplification and moving average calculations to obtain parameters defining the linear curve segments of the selected power intervals. Means, means for calculating a gain factor through parameters defining a linear curve segment of the selected power interval, and means for obtaining a predistortion signal.

According to another aspect of the present invention, there is provided a transmitter of a communication system, the transmitter comprising means for determining the power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Means for obtaining parameters defining a linear curve segment of the selected power interval, means for calculating a gain factor through the parameters defining a linear curve segment of the selected power interval, and means for obtaining a predistortion signal. It features.

According to another aspect of the present invention, there is provided a receiver of a communication system, the receiver defining means for defining in-phase and quadrature differences of symbol determination and received signal samples for each constellation point, and the defined above. And transmission means for transmitting in-phase and quadrature differences to the transmitter.

According to another aspect of the invention, there is provided a transmitter of a communication system, the transmitter determining power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval. Means for averaging the means, the in-phase and orthogonal differences, rotating the in-phase and orthogonal differences averaged toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations to define a linear curved segment of the selected power interval. Average means for obtaining the data, a calculation means for calculating a gain factor through parameters defining a linear curve segment of the selected power interval, and an acquisition means for obtaining a predistortion signal.

According to another aspect of the invention, there is provided a transmitter of a communication system, the transmitter determining power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval. Means for determining parameters defining a linear curve segment of the selected power interval, calculation means for calculating a gain factor through the parameters defining a linear curve segment of the selected power interval, and obtaining a predistortion signal. It characterized in that it comprises an acquisition means.

According to another embodiment of the present invention, there is provided a predistorter measuring device of a receiver, which predetermines the in-phase and quadrature differences of symbol determination and received signal samples for each constellation point. Means, and means for transmitting the in-phase and quadrature differences defined above to a transmitter.

According to another embodiment of the present invention, there is provided a predistorter structure of a transmitter, the predistorter structure determining the power of a signal to be predistorted, quantizing the power in a predetermined manner and corresponding power spacing. Power processing means to select a mean, the in-phase and orthogonal differences are averaged, rotate the averaged in-phase and orthogonal differences toward in-phase axes of the in-phase and orthogonal planes, and perform amplification and moving average calculations to perform linear curve segments of the selected power intervals. An average means for obtaining parameters defining a P, a calculation means for calculating a gain factor through parameters defining a linear curve segment of the selected power interval, and an acquisition means for obtaining a predistortion signal.

According to another embodiment of the present invention, there is provided a predistorter structure of a transmitter, the predistorter structure determining the power of a signal to be predistorted, quantizing the power in a predetermined manner and corresponding power spacing. Power processing means for selecting a; determining means for determining parameters defining a linear curve segment of the selected power interval; calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected power interval; And acquiring means for obtaining a predistortion signal.

According to another aspect of the present invention, there is provided a connection device of a communication system, the connection device determining power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval. Power processing means, average the in-phase and orthogonal differences, rotate the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and perform amplification and moving average calculations to determine the linear curve segments of the selected corresponding power intervals. An average means for obtaining defining parameters, a calculation means for calculating a gain factor through parameters defining a linear curve segment of the selected power interval, and an acquisition means for obtaining a predistortion signal.

According to another aspect of the present invention, there is provided a connection device of a communication system, the connection device determining power of a signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval. Power processing means, determining means for determining parameters defining a linear curved segment of the selected power interval, calculating means for calculating a gain factor through parameters defining a linear curved segment of the selected power interval, and a predistortion signal Characterized in that it comprises a means for obtaining.

Embodiments of the invention are described in the dependent claims.

The method and system of the present invention provide several advantages. In a preferred embodiment of the present invention, a power amplifier predistorter is provided that reduces signal spectral spread.

Hereinafter, the present invention will be described in more detail with reference to preferred embodiments and the accompanying drawings.

1 is a flowchart.

2A shows an amplitude interval.

가 어떠한 방식으로 진폭 이득과 동일해지는지를 보여주는 도면이다.2b shows one curved segment, i.e.

Is a diagram showing how is equal to the amplitude gain.

3 shows a receiver.

4 shows a transmitter.

5 shows a microprocessor of transmission.

,

)를 통한 사전­왜곡기 이득(

)의 실수부의 정의를 분명하게 하기 위한 한 곡선 세그먼트를 보여주는 도면이다.6 is a coefficient (

,

Predistorter gain ()

Figure 1 shows a curved segment to clarify the definition of the real part of.

1 is a flowchart illustrating one embodiment of a pre-distortion method performed in a telecommunications system in which the telecommunications system includes at least one transmitter and at least one receiver providing receiver-side distortion measurements. Usually, the system also includes means for transmitting distortion measurements from the receiver to the transmitter, because the receiver and the transmitter are typically not located adjacent to each other. For example, the transmitter of the system uses an adaptive discriminating linear method. Typically, measurements and pre-distortions are made for each symbol. The main purpose of the pre-distortion is to be able to use high transmit power in the non-linear range of the power amplifier without violating the spectral mask.

Since the measurements are typically made at the receiver and the actual predistortion is made at the transmitter, there is data to be exchanged between the communication elements. Data exchange is typically done via a so-called back channel used for link management data exchange.

The implementation of the method may vary, where in-phase (in-phase) and Q (quadrature) difference measurements are preferably made at the receiver and the actual signal predistortion is made at the transmitter, but other method steps ) May be made on both sides. Typically, the actual predistortion steps are performed at the transmitting side, since the transmitter has much knowledge of the most suitable parameters during the adaptation process.

The embodiment begins at block 100. At block 102, in-phase (I) and quadrature (Q) differences between symbol determination and received signal samples for each constellation point are typically defined at the receiver and transmitted to the transmitter.

The definition is made at each sampling interval. The term 'I and Q difference' means the same as 'I and Q error' (when I means in phase and Q means orthogonal).

The I and Q values should be measured separately for each possible magnitude (where 'mgrnitude' refers to a circle of constellation points located approximately equal distance from the origin in the constellation diagram). it means). When multistage-QAM modulation, such as 32 quadrature amplitude modulation (QAM) is used, for example 32 QAM, the number of measurements by converting 32 constellation points into the first quadrant of the constellation diagram circle It may be possible to minimize the number to eight times. The average for each measurement is typically calculated by accumulating a programmable number of measurements.

The following embodiment steps are typically performed in the predistorter of the transmitter. The transmitter's pre-distorter may be located at the base station (also referred to as Node B) or at an equivalent network component of the communication system. The system may also be a point-to-point (or point-to-point, multi-point-to-point, mesh-wireless network, etc.) system, in which case the predistorter may be located in a component that contains the equipment needed. have. The network component is referred to herein as a connection device.

In the predistortion performed at the transmitter, it is important to consider the data units of the difference measurements, the receiver measuring and returning the measurement data using specific data units, and the data format used is adapted for the interpretation of the data. The transmitter must know.

In block 104, power intervals are formed based on the modulation methods used. Amplitude intervals (obtained from amplitude values by power values squared) are described in more detail with the aid of the example shown in FIG. 2A. It is typical for power intervals to be defined before initiating the predistortion process for each modulation method used in the communication system. It is possible to update the power intervals during the predistortion process but in practice this is typically not necessary. In FIG. 1, the arrow 116 shows the possibility of forming power intervals only once.

In FIG. 2A, the amplitude interval is shown. On the x-axis 200, there are input amplitude values and on the Y-axis 202, there are output amplitude values. The example of FIG. 2A shows a 32 QAM system in which there are five different sizes representing the amplitudes of the constellation points represented by points 204, 206, 208, 210 and 212. As can be seen in FIG. 2A, the amplitude gap curve 214 can be considered to consist of several straight lines (curved segments). In FIG. 2A the curve segments are labeled # 1, # 2, # 3, # 4. Thus, the curve 214 can be thought of as a piecewise linear.

이다.In digital hardware, it is easier to calculate the power than the amplitude of complex values, thereby making it easier to determine which input power interval the current input value belongs to rather than which amplitude interval. The corresponding input power to output power relationship is very similar to the input amplitude to output amplitude relationship shown in FIG. 2A, with only the slope squared. In other words,

to be.

In block 106, the power of the signal to be pre-distorted is determined, and the power is quantized in a predetermined manner and a straight curve segment to which the corresponding power interval, e.g., the current input value belongs, is selected. As mentioned above, in digital hardware, it is easier to calculate the power than the amplitude of the complex value, thereby determining which input power interval the current input value belongs to rather than which amplitude interval. It is easier to do.

In the transmitter, it is desirable to place the predistorter after digital pulse shaping and low pass filters so that the signal spectrum can be formed to meet the requirements of the system specification. However, digital pulse shaping and low pass filters produce additional values due to interpolation between constellation points. The complex amplitude of such values may be above or below the complex amplitude of the constellation points, depending on the modulation method used. The complex amplitude may also be between two constellation points. The receiver provides information about in-phase (I) and quadrature (Q) component differences only for constellation points. As the amount of pre-distortion I and Q (corresponding to amplitude and phase) compensation amount depends on the complex input amplitude, an output value is also generated for values different from the actual constellation points.

For values different from the constellation points (reference points), interpolation or extrapolation is used for approximation, with interpolation for the intermediate values (middle reference points) and the outermost constellation. Extrapolation is used for values (reference points) with complex amplitudes above the circle or below the innermost circle.

Linear interpolation (or extrapolation) is chosen in this embodiment due to the simplicity, but other methods may also be used. There are several possibilities, for example, to increase the number of linear curve segments that make the curve of the input and output relationships smooth, and / or to use higher order interpolation methods such as cubic interpolation. In addition, the higher order interpolation method can be used as an intermediate step such that the desired pre-distortion curves are first calculated using higher order interpolation / extrapolation methods and then the desired curves are approximated by using multiple distinct linear segments.

In FIG. 2A, an example is shown of 32 QAM systems in which there are five different magnitudes representing the amplitudes of the constellation points represented by points 204, 206, 208, 210, and 212.

There are three different cases, first the amplitudes between the two circles of the constellation diagram are processed. In this embodiment, linear interpolation between the constellation points in question is used, for example, between the constellation points 208 and 210. Second, the amplitudes below the innermost circle are processed. In this embodiment, the extrapolation of the linear interpolation line between the two innermost constellation circles 210, 212 extends toward the y-axis indicated by line 216. An alternative embodiment is to interpolate between the innermost circle and the origin. Third, the amplitudes on the outermost circle are processed. In the case of extrapolation, a linear interpolation line between two outermost circles 204 and 206 extends, indicated by line 218.

Power is quantized for input power intervals defined by the modulation method used. In the example of FIG. 2A, that is to say 32 QAM, there are four power intervals so that power is quantized according to these levels. At this time, the linear curve segment to which the corresponding power interval, for example the current input value belongs, is selected.

In the following, complex values are represented using underscores.

Predistortion has the following form in Cartesian coordinate system.

In the above formula,

는 사전­왜곡기의 복소 입력 신호이며,

Is the complex input signal of the predistorter,

는 사전­왜곡기의 복소 출력 신호이고,

Is the complex output signal of the predistorter,

는 입력 전력(P)에 따른 복소 선형 내삽이다.

Is a complex linear interpolation according to the input power P.

이 사용된 경우보다 적은 계산 노력이 필요하기 때문이다.In Equation 1, power is used because the input amplitude is

This is because less computational effort is required than when used.

At block 108, parameters are defined that define a linear curve segment of the power interval. The angle coefficient (tilt) and y-axis intersection of each of the complex value line segments can be obtained. Since a segmented linear approximation is used, each segment can be written as follows.

In the above formula,

는 입력 전력(

)에 따른 복소 선형 내삽이고,

Is the input power (

Complex linear interpolation according to

는 입력 전력이며,

Is the input power,

은 기울기이고,

Is the slope,

은 y-축 교점이다.

Is the y-axis intersection.

,

)은 본원에서 전력 간격의 선형(복소) 곡선 세그먼트를 정의하는 매개변수들이라고 언급된다.Parameters (

,

) Are referred to herein as parameters defining a linear (complex) curve segment of the power interval.

Next is an example of the calculation of a parameter defining power intervals for 32 QAM. The accumulated I and Q differences transmitted from the receiver to the transmitter are averaged by dividing the accumulated I and Q differences by the number of samples. Averages are made for the I and Q components.

이라고 언급된다. 복소 회전자를 통한 회전은 각각 평균이 이루어진 차의 복소 곱을 통해 이루어진다.If several constellation points are located on the same circle of the constellation diagram, it is possible to calculate an average value for the amplitude and phase difference of such constellation points. In the 32 QAM system, using the possibility to calculate the mean values, it is possible to reduce the number of constellation points to be processed five times. For that purpose, the averaged I and Q differences are rotated toward the in-phase axis of the I and Q planes by the same angle as needed to rotate the corresponding reference point toward the in-phase axis. The resulting complex values are

It is mentioned. Rotation through the complex rotor is achieved through the complex product of the averaged difference respectively.

In order to control the stability of the closed loop, amplification and moving average calculations are made. The purpose of the amplification is to use only part of the measurement errors for pre-distortion.

In principle, the predistorter is designed to minimize the average phase and amplitude errors measured at the receiver's determining device. This target is not advantageous for optimizing the output spectrum of the transmitter but rather for optimizing bit error performance. Fully minimizing phase and amplitude errors (also referred to herein as 'phase and amplitude difference') at the receiver degrades the transmission signal spectrally at high output power levels. Although the in-band spectral distortions are reduced, the out-of-band spectral components can increase so much that the spectral mask is violated.

It is therefore common to trade-off between the bit error performance at the receiver and the signal spectral shape requirements of the transmitted signal. Programmable amplification factor C is designed for such trade-offs. If the factor C is large, the portion of the measurement distortion used to calculate the pre-distortion parameters becomes large.

The moving average has a smoothing effect that can reach steady state from large initial deviations or occasionally cope with unreliable measurements. The smoothing and convergence rate is influenced by the ratio of the factor C and independently programmable factor N. C is used to control the amount of distortion correction, while N is used to control the smoothing and convergence speed.

The amplification and moving average calculations are made for each of the five difference values using the following equation.

In the above formula,

는 5개의 남아 있는 복소값 차 중 하나이고,

Is one of the five remaining complex values,

는 증폭 인자이며,

Is an amplification factor,

는 복소 이동 평균값이고,

Is the complex moving average,

는 원 인덱스(circle index)이며,

Is the circle index,

은 시간 인덱스이고,

Is the time index,

은 반복 이동 평균 계산의 시상수이고,

Is the time constant for iterative moving average calculation,

In addition, the following equation is used.

In the above formula,

는 5개의 남아 있는 복소값 차 중 하나이고,

Is one of the five remaining complex values,

는 증폭 인자이며,

Is an amplification factor,

는 복소 이동 평균값이고,

Is the complex moving average,

는 원 인덱스이며,

Is the raw index,

은 시간 인덱스이고,

Is the time index,

은 반복 이동 평균 계산의 시상수이다.

Is the time constant of the repeated moving average calculation.

를 통해 이동 평균을 계산한 다음에 그 결과를

만큼 증폭하는 것과 같은 다른 계산 방법들이 또한 가능하다.

To calculate a moving average,

Other calculation methods, such as multiplying by as much as possible, are also possible.

은 정상 이동 평균에 이르는데 필요한 시간 및 평활을 정의한다. 고정된

및

값들에 대해, 수렴 및 트래킹(tracking) 속도가 빠르며, 큰

값들에 대해, 수렴 및 트래킹 속도가 낮다.

Defines the time and smoothing required to reach the normal moving average. Fixed

And

For values, the convergence and tracking speed is fast and large

For values, the convergence and tracking speed is low.

이다. 원하는 송신 전력이 비-선형 범위 내에 있는 경우에, 상기 송신 전력은 최종 전력에 이를 때까지 사전-왜곡 매개변수들의 충분한(예컨대, 30) 수렴 라운드(convergence round)를 대기한 다음에 점차로(예컨대, 0.5 dB씩) 증가될 수 있다. 신속한 전력 변화들을 가능하게 하려면, 중간 사전-왜곡 매개변수들 및 대응하는 이동 평균들은 비-선형 범위의 송신 전력 값들에 대한 사전-왜곡 매개변수들 및 대응하는 이동 평균들의 신속한 조회(recall)를 허용하는 룩-업 테이블(look-up-table; LUT)에 저장되는 것이 전형적이다. 이러한 룩-업 테이블(LUT)은 또한 전력 차단(power failure) 이후에도 재사 용할 수 있도록 비-휘발성 메모리에 유지된다.In order to prevent spectral mask violations without predistortion, the initial state of the moving average at start-up at low output powers outside of the non-linear range is typically set to zero, that is to say

to be. If the desired transmit power is within the non-linear range, the transmit power waits for a sufficient (eg, 30) convergence round of pre-distortion parameters until the final power is reached and then gradually (eg, In 0.5 dB increments). To enable rapid power changes, the intermediate pre-distortion parameters and corresponding moving averages allow for a quick recall of the pre-distortion parameters and corresponding moving averages for the non-linear range of transmit power values. Typically stored in a look-up-table (LUT). This look-up table (LUT) is also maintained in non-volatile memory for reuse after power failure.

The amplitude difference and phase difference are then calculated based on the pre-distortion parameters.

)은 다음과 같이 근사화될 수 있다.Since the phase difference can be assumed to be relatively small relative to the amplitude difference, the received amplitude of the particular constellation diagram circle

) Can be approximated as

In the above formula,

는 콘스텔레이션 지점이고,

Is the constellation point,

는 원의 개수를 의미하며,

Means the number of circles,

는 실수부를 의미하고,

Means real part,

는 절대값을 의미하며,

Means absolute value,

은 이동 평균을 의미하고,

Means moving average,

)은 다음과 같이 계산될 수 있다.Receive phase of a particular constellation diagram circle averaged (

) Can be calculated as

In the above formula,

은 허수부를 의미하고,

Means imaginary part,

는 절대값을 의미하며,

Means absolute value,

는 콘스텔레이션 지점이고,

Is the constellation point,

는 실수부를 의미하고,

Means real part,

는 원의 개수를 의미하며,

Means the number of circles,

은 이동 평균을 의미한다.

Means moving average.

)이

보다 작을 경우에, 역 이득차는 1보다 크다. 다시 말하면, 상기 사전-왜곡기는 진폭을 증가시킨다.The pre-distorter is designed to compensate for amplitude and phase distortions. Amplitude compensation may be achieved by dividing the reference amplitudes by the averaged received amplitudes calculated according to equation (5). The received amplitude of the average (

)this

If smaller, the reverse gain difference is greater than one. In other words, the pre-distorter increases the amplitude.

The inverse gain difference is a reception phase in which the average calculated by using Equation 6 × (−1) is obtained.

Therefore, the following equation is obtained.

In the above formula,

는 절대값을 의미하며,

Means absolute value,

는 원(

) 상의 복소 기준 지점이고,

Is the circle (

Is the complex reference point on

는 수학식 5로부터 구해지고,

Is obtained from Equation 5,

In the above formula,

는 수학식 6에서 구해진다.

Is obtained from equation (6).

The above equations are obtained for a coordinate system utilizing polar coordinates. Since it is common for pre-distorters to operate in Cartesian coordinate systems, Equations 7 and 8 should be replaced with Cartesian coordinates. Therefore, the following equation is obtained for the real part of the complex correction coefficient,

For the imaginary part, the following equation is obtained.

Note that Equations 9 and 10 are valid only for constellation points, not for intermediate points. The correction factors for the intermediate points can be obtained by interpolation as shown above. Sine and cosine functions can be made simpler by using truncated Taylor series approximation. Taylor feedwater is well known to those skilled in the art.

In case the pre-distortion parameters are stored in hardware HW, the resolution is limited. The parameters are rounded to the allowed bit size, optionally noting that linear interpolations do not lead to jumps of pre-distortion gain parameters at the boundaries of the segments.

, 즉 복소 이득의 절대값은

및

간의 이득 인자와 등가이다.In Figures 2A and 2B, the y-axis shows output amplitude values.

, That is, the absolute value of the complex gain

And

It is equivalent to the gain factor of the liver.

)은 다음과 같이 표기될 수 있다.Power of reference circles (

) Can be written as

In the above formula,

는 절대값을 의미하며,

Means absolute value,

는 원(

) 상의 복소 기준 지점이다.

Is the circle (

) Is the complex reference point on.

The y-axis intersection (output amplitude or power) and angle coefficient for each complex line segment can be found as follows when the graph is considered to be a discrete linear.

,

)의 실수부들에 대한 정의를 명확하게 하기 위하여, 도 6에 한 곡선 세그먼트(604)가 도시된 것이다. 수직축(600) 상에는 이득 인자(

)의 실수부가 존재하며 수평축(602) 상에는 입력 전력(

)이 존재한다. 사전-왜곡 방법에서는, 복소 이득 인자(

)의 y-축 교차점(610; (

)) 및 기울기(

)를 나타내는 복소(

,

) 인 자들은 수학식 9 및 수학식 10에 따라 구해진다.One curved segment 604 is shown in FIG. Pre-distorter coefficients (

,

In order to clarify the definition of the real parts of), one curved segment 604 is shown in FIG. 6. The gain factor (on the vertical axis 600)

) Real part and on the horizontal axis 602 the input power (

) Exists. In the pre-distortion method, the complex gain factor (

Y-axis intersection point 610;

)) And slope (

Complex ()

,

) Factors are obtained according to equations (9) and (10).

)의 절대값과 동일한 진폭 이득 인자만큼 입력 진폭에 관련된다. 출력 위상은 복소 이득 인자(

)의 위상과 동일한 위상 이득 인자만큼 입력 위상에 관련된다.The output amplitude (see FIGS. 2A and 2B) is a complex gain factor (

The input amplitude is related by an amplitude gain factor equal to the absolute value of. The output phase is a complex gain factor (

Is related to the input phase by a phase gain factor equal to that of.

),

)의 입력 전력들(

)이 도시되어 있다. 진폭 차(

; 608)는 수직축에서 구해지며 입력 전력 차(

; 606)는 수평축에서 구해진다. 기울기(

)의 실수부가

이다.Two consecutive complex reference points (

),

Input powers of

) Is shown. Amplitude difference (

; 608 is obtained from the vertical axis and the input power difference (

; 606 is obtained from the horizontal axis. inclination(

Real part of)

to be.

및

을 보여주기 위해 구성 라인으로서 점선들이 도 2b에서 사용된다.

,

및

의 허수부들에 대해 유사한 도면이 도시될 수 있다.

And

Dotted lines are used in FIG. 2B as configuration lines to show

,

And

A similar figure can be shown for the imaginary parts of.

Each segment is

As shown above in Equation 2,

In the above formula,

는 입력 전력(

) 또는 이득 인자에 따른 복소 선형 내삽이고,

Is the input power (

) Or complex linear interpolation depending on the gain factor,

는 입력 전력(또는 진폭)이며,

Is the input power (or amplitude),

은 각도 계수(기울기)이고,

Is the angle factor (tilt),

은 y-축 교점이다.

Is the y-axis intersection.

및

(전력 간격의 선형 곡선 세그먼트를 정의하는 매개변수들)의 계산이 더 상세하게 설명된다.In the following,

And

The calculation of (parameters defining the linear curve segment of the power interval) is described in more detail.

의 실수부 및 허수부가 다음과 같이 계산된다.Since a 32 QAM system is used herein as an example, the following equations are written for four line segments from five complex gain values corresponding to five reference circles. A new index l (l is given from 1 to 4) is employed to mean each line segment. first,

The real part and the imaginary part of are calculated as follows.

In the above formula,

은 실수값을 의미하며,

Means a real value,

는 입력 진폭이고,

Is the input amplitude,

는 절대값을 의미하며,

Means absolute value,

는 당해 원 상의 복소 기준 지점을 의미하고(펄스 샤프닝(pulse sharpening)과 같은 변조기 및 사전-왜곡기 간의 가능한 회로 증폭을 포함함),

Means the complex reference point on the circle (including possible circuit amplification between the modulator and pre-distorter, such as pulse sharpening),

In the above formula,

는 허수값을 의미하며,

Means an imaginary value,

는 입력 진폭이고,

Is the input amplitude,

는 절대값을 의미하며,

Means absolute value,

는 당해 원 상의 복소 기준 지점을 의미한다(펄스 샤프닝(pulse sharpening)과 같은 변조기 및 사전-왜곡기 간의 가능한 회로 증폭을 포함함).

Means the complex reference point on the circle (including possible circuit amplification between the modulator and pre-distorter, such as pulse sharpening).

의 실수부 및 허수부가 다음과 같이 계산된다.next time

The real part and the imaginary part of are calculated as follows.

In the above formula,

은 실수값을 의미하며,

Means a real value,

는 입력 진폭이고,

Is the input amplitude,

는 절대값을 의미하며,

Means absolute value,

는 당해 원 상의 복소 기준 지점을 의미하고(펄스 샤프닝과 같은 변조기 및 사전-왜곡기 간의 가능한 회로 증폭을 포함함),

Means the complex reference point on the circle (including possible circuit amplification between the modulator and pre-distorter such as pulse sharpening),

은 수학식 12를 사용함으로써 계산되는 각도 계수의 실수부이다.

Is a real part of the angle coefficient calculated by using Equation (12).

의 허수부는 다음과 같이 계산된다.

The imaginary part of is computed as

In the above formula,

은 허수값을 의미하며,

Means an imaginary value,

는 입력 진폭이고,

Is the input amplitude,

는 절대값을 의미하며,

Means absolute value,

는 당해 원 상의 복소 기준 지점을 의미하고(펄스 샤프닝과 같은 변조기 및 사전-왜곡기 간의 가능한 회로 증폭을 포함함),

Means the complex reference point on the circle (including possible circuit amplification between the modulator and pre-distorter such as pulse sharpening),

은 수학식 13을 사용함으로써 계산되는 각도 계수의 허수부이다.

Is an imaginary part of the angle coefficient calculated by using equation (13).

)가 당해 전력 간격의 복소 선형 곡선 세그먼트를 정의하는 매개변수들, 즉

및

(수학식 2 참조)를 사용함으로써 계산된다. 신호는 이 신호에 상기 이득 인자(

)를 곱함으로써 사전-왜곡된다.In block 110, having a complex value form a general gain factor (

) Define the complex linear curve segments of the power interval, i.e.

And

(See Equation 2). Signal is the gain factor (

Pre-distorted by multiplying

The embodiment ends at block 112. Arrow 114 shows one possibility for repeating the above embodiment. It is common for the method to require several iterations to converge. Due to temperature changes and aging, updating of such pre-distortion parameters may be necessary.

The method may be implemented in various types of communication systems. One example is point to point (or one point to many point) systems. For point-to-point (one-to-many) systems, it is possible to apply hot standby (HSB) device protection methods. In the hot standby method, there are one or more secondary radio units that are prepared to adopt the functions of the active radio unit in the event of a failure. There are several types of failures such as the wireless unit becomes unavailable, or there is a degradation in signal quality that causes the reception of the wireless signal to be not continuous or causes too many bit errors upon detection. The name "standby" derives from the fact that the transceivers of both radio units are switched on but the guard radio unit is in "hot standby" mode because the transmitters of the guard radio unit are muted.

=0 및

=1일 수 있다. 이후의 프로세스는 다음과 같은 반복 단계들에 대해 측정 결과들을 사용하여 속행된다.The procedure of defining the pre-distortion parameters can be described as two nested loops because the outer loop increases the transmit power in small steps until the target transmit power is reached. The inner loop performs pre-distortion updates for fixed transmit power starting from the minimum power that does not cause a spectral mask violation. The default parameters for the pre-distorter on first start up are for example

= 0 and

May be = 1. The subsequent process is continued using the measurement results for the following repeating steps.

It is common for the transmit power not to increase above a predetermined limit to prevent spectral mask violations. In the event of reaching typical implementation dependent limits set for pre-distortion, the microprocessor limits the increase in transmit power. Such a restriction also prevents internal overflow in addition to preventing spectral mask violations.

In addition, radio conditions can change quite rapidly, in which case the system must switch quickly from one transmit power to another. In order to avoid multiple iteration steps in which each iteration has a small power change, pre-calculated pre-distortion parameters and moving averages can be stored in the look-up table (LUT). The look-up table (LUT) may also be stored in permanent and non-volatile memory to allow the use of parameters after power off or at new startup. In addition, regular updates of pre-distortion parameters can be made to adjust the system to the effects of aging or temperature change. The sequence number can be used to avoid using the same measurement information twice.

In the following an example of the devices used for pre-distortion measurements at the receiver is described with the aid of FIG. 3. The device is typically located near the judging device as can be seen in FIG. The receiver may for example be located at a user terminal of a communication system. The receiver may also be located at a base station (also referred to as Node B) or at an equivalent network component of the communication system. The system may also be a point-to-point (or point-to-point, multi-point-to-point, mesh-wireless network, etc.) system, in which case it may be located in a component that includes facilities requiring pre-distorters. have. The network component is referred to herein as a connection device.

It will be apparent to those skilled in the art that the receiver may also include other components than shown in FIG.

In this embodiment, the received signal sample arrives at the judging device 300. The received signal sample and detected symbol are employed in rotation block 302. In the rotating block, the sample and the detected symbol are rotated in the first quadrant of the constellation diagram to minimize the number of measurements needed. Constellation diagrams are known to those skilled in the art.

The difference (error) between the received signal sample and the detected symbol is calculated for in-phase and quadrature components at block 306. The in-phase component difference is the difference between the in-phase component of the received sample and the in-phase component of the detection symbol. Thus, the orthogonal component difference is the difference between the orthogonal component of the received signal sample and the orthogonal component of the detected symbol.

Usually, amplitude and phase differences are calculated in the orthogonal region by calculating the I and Q differences.

Blocks 304 and 308 supply the calculated differences to corresponding memory units 310A-H (only the first and last memory units are shown in FIG. 3) according to the constellation point. When 32 QAM is used and the signal samples along with the detected symbols are rotated into the first quadrant, eight different constellation points, resulting in 8 * 2 = 16 (individual storage for real and imaginary values). Devices) or portions of a memory unit).

The microprocessor 314 may average the differences to improve the accuracy of the amplitude and phase information sent to the transmitter. It is also possible to perform averaging at the transmitter, in which case the main task of the microprocessor is to control the supply of information to the transmitter. Block 312 is a control block that controls the memory usage and computation process in the microprocessor 314. The above-mentioned device is typically located in an application-specific integrated circuit (ASIC) component. Other possible implementations are field-programmable gate arrays (FPGAs), digital signal processing (DSP) or even general purpose processors. The required processing speed largely determines the implementation.

In the following, a generalized example of the pre-distorter of the transmitter is described in more detail with the aid of FIGS. 4 and 5. It will be apparent to those skilled in the art that the pre-distorter may also include other components than those illustrated in FIGS. 4 and 5. The pre-distorter may be located in an equivalent network component of a base station (also referred to as Node B), a radio network controller, or a communication system. The system may also be a point-to-point (or point-to-point, multi-point-to-point, mesh-wireless network, etc.) system, in which case the pre-distorter may be located in a component containing the installations needed. Can be. The network component is referred to herein as a connection device.

The pre-distorter is usually located after the digital pulse shaping and low pass filters if the signal spectrum should be improved. However, these filters produce signal values between constellation points. Therefore, interpolation and / or extrapolation are used as described above.

If the harmonics of the baseband signal should be compensated for, over-sampling is typically needed. If a sample rate equal to or higher than four times the symbol rate is used, then the over-sampling requirements are typically met. The spectral components produced by the pre-distorter are again aliased in the frequency domain up to the Nyquist frequency. Therefore, if the sample rate is high, the pre-distortion result is good because the spectral components with high frequencies attenuate more quickly than the components with low frequencies. Another advantage of using very high over-sampling is that depending on the absolute value of the over-sampling factor, compensation up to harmonics higher than the third harmonic of the baseband spectrum is also possible.

The input signal of the pre-distorter is sent to block 400, where the signal power is determined. The determined power is sent to block 402, where the power values are quantized based on the power intervals used in the present embodiment. In the case of 32 QAM, there are four different power intervals. 2 shows an example of the power interval. At block 404, it is determined to which power interval (linear curve segment) the current input value belongs.

및

를 계산하다. 그러한 계산은 도 5 및 도 2b의 도움으로 설명된다. 도 2b에는, 한 곡선 세그먼트의 기울기, 다시 말하면

가 어떠한 방식으로 진폭 이득과 동일해지는지가 도시되어 있다. 점선은 구성 라인으로서 사용된다.Microprocessor 406 is responsible for the parameters defining complex linear curve segments of the power intervals, i.e.

And

Calculate Such a calculation is explained with the aid of FIGS. 5 and 2B. 2b shows the slope of one curved segment, ie

Is shown how is equal to the amplitude gain. Dotted lines are used as construction lines.

The receiver transmits I and Q information to the transmitter on the back channel. At the transmitter, the information is sent to the microprocessor 406. In this example, there are eight different input values.

In block 500, the I and Q differences are averaged (if this is not done at the receiver). The averaged differences are then rotated toward the in-phase axis at block 502. The angle of rotation is typically the angle at which it is necessary to rotate the corresponding reference point towards the I axis. Such rotation can be achieved by complex multiplying the average of the respective complex rotors for each corresponding reference point.

At block 504, rotated differences belonging to the same constellation circle are averaged. When 32 QAM is used, it is possible to form the model shown in FIG. 2A and there are five different output values from block 504. In block 506, amplification and moving average calculations are made to identify a stable pre-distorter where no overcompensating is performed.

및

가 계산된다. 그러한 계산은 위에서 상세하게 설명되었다.In block 508, parameters defining a complex linear curve segment of power intervals, i.e.

And

Is calculated. Such calculations have been described in detail above.

,

)이 블록(408)에서 선택된다. 콘스텔레이션 지점들과는 다른 값들에 대해, 블록(410)에서 근사를 위한 내삽 또는 외삽이 사용되는데, 중간 값들에 대하여는 내삽이 사용되고 최외곽 콘스텔레이션 원 위에 있거나 최내곽 원 아래에 있는 진폭들을 지니는 값에 대하여는 외삽이 사용된다. 선형 내삽(또는 외삽)은 간략성을 위해 본 실시예에서 사용되지만, 다른 방법들이 또한 사용될 수 있다.Parameters corresponding to the current line segment of the power diagram (see FIG. 6) (

,

Is selected at block 408. For values other than the constellation points, an interpolation or extrapolation for approximation is used at block 410, with interpolation being used for intermediate values and values with amplitudes above or below the outermost constellation circle. Extrapolation is used for. Linear interpolation (or extrapolation) is used in this embodiment for simplicity, but other methods may also be used.

)가 결정된다.In block 410, the gain factor (

) Is determined.

)에 의해 곱해진다.In order to obtain a complex predistortion signal, the complex input signal of the pre-distorter is converted into a complex gain factor (block 412).

Is multiplied by

The above mentioned devices are typically located in one or more application specific integrated circuit (ASIC) components. Other possible implementations are field programmable gate arrays (FPGA), digital signal processing (DSP) or even general purpose processors. The required processing speed largely determines the implementation.

The implementations may vary. It is preferred that I (in-phase) and Q (orthogonal) difference measurements are performed at the receiver, while the actual signal pre-distortion is performed at the transmitter, but other procedural steps can be made at both sides.

While the invention has been described above with reference to examples in accordance with the accompanying drawings, it is apparent that the invention is not limited to such examples and may be modified in many ways within the scope of the appended claims.

Claims (31)

A pre-distortion method in a communication system, the pre-distortion method in which the communication system includes at least one transmitter and at least one receiver, The pre-distortion method, Defining in-phase and quadrature differences of the symbol determination and received signal sample for each constellation point, and transmitting the defined in-phase and orthogonal differences to a transmitter; Forming power intervals based on the modulation method used; Determining a power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; A parameter that defines the linear curve segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculation Obtaining them; And Calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval to obtain a predistortion signal. A pre-distortion method performed in a communication system, wherein the communication system includes at least one transmitter and at least one receiver, Pre-distortion method Defining in-phase and quadrature differences of the symbol determination and received signal sample for each constellation point, and transmitting the defined in-phase and orthogonal differences to a transmitter; Forming power intervals based on the modulation method used; Determining a power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Obtaining parameters defining a linear curve segment of the selected corresponding power interval; And Calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval to obtain a predistortion signal. The method of claim 2, The pre-distortion method, And averaging the in-phase and orthogonal differences directly at in-phase and orthogonal levels. The method of claim 2, The pre-distortion method, Minimizing the number of measurements to define in-phase and quadrature differences of the received signal sample by rotating the constellation points in the first quadrant of the constellation diagram. The method of claim 2, The pre-distortion method, Using interpolation to form a gain factor for the reference points between the two constellation diagram circles. The method of claim 2, The pre-distortion method, Using extrapolation to form gain factors for reference points above or below the outermost circle of the constellation diagram. The method of claim 2, The pre-distortion method, And averaging distortion measurements for reference points on the same constellation circle. The method of claim 2, And said forming step comprises forming power intervals based on the modulation method used, wherein the modulation method used in said communication system comprises multi-stage orthogonal amplitude modulation. The method of claim 1, The pre-distortion method,

And amplifying and moving average calculation as in the previous step. The method of claim 1, The pre-distortion method,

And amplifying and moving average calculation as in the previous step. The method of claim 2, The pre-distortion method, Mediating the linearly curved segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations. And further obtaining the variables. In the receiver of a communication system, The receiver, Means for defining in-phase and quadrature differences in symbol determination and received signal samples for each constellation point; And Means for transmitting said defined in-phase and quadrature differences to a transmitter. The method of claim 12, The receiver, And means for minimizing the number of measurements to define in-phase and quadrature differences of the received signal sample by rotating the constellation points in the first quadrant of the constellation diagram. 13. The receiver of claim 12, wherein the modulation method used in the receiver comprises multistage quadrature amplitude modulation. In the transmitter of a communication system, The transmitter, Means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; A parameter that defines the linear curve segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations Means for obtaining them; Means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And Means for obtaining a pre-distortion signal. In the transmitter of a communication system, The transmitter, Means for determining the power of the signal to predistort, quantize the power in a predetermined manner and selecting a corresponding power interval; Means for obtaining parameters defining a linear curve segment of the selected corresponding power interval; Means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And Means for obtaining a pre-distortion signal. The method of claim 15, The transmitter, And means for forming a gain factor for reference points between two constellation diagram circles using interpolation. The method of claim 15, The transmitter, And means for forming a gain factor for reference points that are above or below the outermost circle of the constellation diagram using extrapolation. The method of claim 15, The transmitter, And means for averaging reference points on the same constellation circle. The method of claim 15, The modulation method used in the communication system comprises multi-stage quadrature amplitude modulation. The method of claim 15, The transmitter,

And means for performing amplification and moving average calculations as follows. The method of claim 15, The transmitter of the communication system,

And means for performing amplification and moving average calculations as follows. The method of claim 16, The transmitter, Mediating the linearly curved segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations. And means for obtaining the variables. In the receiver of a communication system, The receiver, Defining means for defining a symbol determination and in-phase and quadrature differences of a received signal sample for each constellation point; And And transmitting means for transmitting the in-phase and quadrature differences defined above to a transmitter. In the transmitter of a communication system, The transmitter, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; A parameter that defines the linear curve segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations Average means for obtaining them; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And means for obtaining the predistortion signal. In the transmitter of a communication system, The transmitter, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Determining means for determining parameters defining a linear curve segment of the selected corresponding power interval; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And means for obtaining the predistortion signal. In the pre-distorter measuring device of the receiver, The pre-distorter measuring device, Means for defining in-phase and quadrature differences in symbol determination and received signal samples for each constellation point; And And means for transmitting the in-phase and quadrature differences defined above to a transmitter. In the predistorter structure of the transmitter, The pre-distorter structure, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; A parameter that defines the linear curve segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations Average means for obtaining them; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And a obtaining means for obtaining a predistortion signal. In the predistorter structure of the transmitter, The pre-distorter structure, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Determining means for determining parameters defining a linear curve segment of the selected corresponding power interval; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And a obtaining means for obtaining a predistortion signal. In the connecting device of the communication system, The connection device, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; A parameter that defines the linear curve segments of the selected corresponding power intervals by averaging the in-phase and quadrature differences, rotating the averaged in-phase and orthogonal differences toward the in-phase axis of the in-phase and orthogonal planes, and performing amplification and moving average calculations Average means for obtaining them; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And obtaining means for obtaining a pre-distortion signal. In the connecting device of the communication system, The connection device, Power processing means for determining the power of the signal to be pre-distorted, quantizing the power in a predetermined manner and selecting a corresponding power interval; Determining means for determining parameters defining a linear curve segment of the selected corresponding power interval; Calculating means for calculating a gain factor through parameters defining a linear curve segment of the selected corresponding power interval; And And obtaining means for obtaining a pre-distortion signal.

Images